PROJECT SUMMARY/ABSTRACT Genomic information is a new form of data that has the potential to more accurately tailor treatments to the individual and improve health outcomes. However, most physicians do not have the necessary training and experience to incorporate this type of data into their practice without assistance. Pharmacogenomic (PGx) clinical decision support (CDS) systems are a new generation of CDS that have the potential to close this gap by incorporating genetic information into electronic health record (EHR) workflows for the first time. As participants in the Electronic Medical Records and Genomics (eMERGE) Network, Northwestern University (NU) has implemented first-generation PGx CDS in the clinical setting. However, outstanding design and implementation questions must be answered before PGx CDS tools can be more broadly disseminated. We do not fully understand how clinicians react to PGx CDS alerts, how they incorporate alert advice into decision- making, or how to best design the tools to ensure compliance and improve care. Preliminary research has shown low physician compliance, indicating that current tools are inadequate. We hypothesize that the application of implementation science (IS) methods to the development of PGx CDS will improve physician adoption and compliance rates. We will utilize a variety of user-centered and socio- technical design methods to address PGx CDS design questions. First, we will conduct an analysis of NU EHR alert log data and clinician interviews to understand how existing first-generation PGx CDS systems are used and perceived. Second, we will conduct targeted interviews and focus groups to more fully understand physician workflows and genomic data needs, from which we will propose modifications to the PGx CDS tools currently in use. Finally, we will follow up with a systematic evaluation of these proposed modifications through prototyping and structured usability testing. This work will result in a set of refined PGx CDS alerts for physicians to apply to patient care, and will also serve as a model for future PGx CDS alert development. The applicant, Tim Herr, is a PhD Candidate in the Health Sciences Integrated PhD Program at the NU Feinberg School of Medicine. His research, beginning at the Master's level, has focused on the development and implementation of PGx CDS. This F31 award would allow Mr. Herr the opportunity to supplement his extensive technical abilities with the additional skills necessary for a career as an independent research scientist. This award would allow him to make additional scientific contributions to the development of PGx CDS while building new expertise in fundamental IS research skills, such as interview techniques and usability testing, and also complete additional training in human-computer interaction.